Embodied Task Planning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
73.5Overall Performance
24
May 19, 2026
84.72SR (Examine&Light)
21
Feb 26, 2026
31.2Success Rate
18
Apr 21, 2026
97.5SP%
17
Jul 1, 2026
97.3Success Rate (SP)
17
Jul 1, 2026
98.7Success Rate (Avg)
17
Feb 26, 2026
97Pass@1 Accuracy
15
Feb 26, 2026
86Pass@1 Accuracy
15
Feb 26, 2026
82.27Examine&Light SR
15
Feb 26, 2026
77.6Avg Success Rate
10
Apr 10, 2026
94Success Rate
9
Apr 10, 2026
92.5Success Rate
9
Apr 10, 2026
1.32TEI
8
Mar 4, 2026
1.36TEI
8
Mar 4, 2026
1.12TEI
8
Mar 4, 2026
97Pass@1 Accuracy
7
Feb 26, 2026
59Success Rate
6
Apr 10, 2026
62.75Success Rate
6
Feb 26, 2026
68.83Success Rate (SR)
6
Feb 26, 2026
80.16Success Rate
6
Feb 26, 2026
71.89Success Rate
6
Feb 26, 2026
72.05Success Rate (SR)
6
Feb 26, 2026
68.4Plan Judge Score
5
Jun 18, 2026
10.42Cumulative Return
5
Jun 11, 2026
82.9Simple Success Rate
4
Feb 26, 2026